Rainfall estimates using the Climate Hazards Group InfraRed Precipitation with Station (CHIRPS) dataset constitute a promising complement to rain gauge networks in areas with fewer stations. To determine their usefulness in practical application, these estimates must be validated and the performance of CHIRPS must be accurately determined. In this study, CHIRPS datasets were validated for a coastal region with extensive plains and complex topography in northern Colombia near the Caribbean Sea. This region presents a complex topography, posing significant challenges for retrieval of rainfall data using remote sensing. Daily, monthly, and annual estimates using CHIRPS were compared with records from 37 rain gauges between 1981 and 2020. Continuous and categorical statistical metrics were applied to evaluate performance in estimating and detecting rainfall by analyzing topographic and climatic constraints. CHIRPS performed best at monthly and annual scales (r > 0.79, NRMSE<1.25, POD>0.87, and FAR<0.36). On daily timescale, its performance tended to be poor to moderate in all rain gauge areas (r < 0.24, NRMSE>5.13, POD>0.39, and FAR<0.84), particularly in coastal regions from January to March. However, during the most intense periods of the El Niño-Southern Oscillation cold phase, the daily timescale performance of CHIRPS improved (NRMSE>3.47, POD>0.45, and FAR<0.64). Overall, it successfully represented spatiotemporal coverage of monthly rainfall averages and is a valuable source of monthly rainfall data making it useful for hydrometeorological modeling in complex topographies with coastal influence.